Stable Diffusion with Viktoria | S02 E14 | Build On Weekly

Playing around with Stable Diffusion on AWS

AWS Admin
Amazon Employee
Published Apr 21, 2023
Last Modified Jun 25, 2024
Viktoria, Jacquie and Darko looking all confused on the live stream
In today's episode Jacquie and Darko are joined by Viktoria. She brings us her experience with starting and playing around with Stable Diffusion on AWS. Generative AI is all the rage now, so Viktoria wanted to see how she can fiddle around with AI Image Generation, for the amazing price of Free 💰🖼️.
On this episode we go over the ways you can get already trained models from our friends over at Hugging Face, load them up in a development environment, and keep on inferencing. The development environment we used is Amazon SageMaker Studio Lab, a free ML learning environment that gives you a bit of compute, a bit of storage and all the security. For the low, low price of FREE 😍.
Initially we tried to use the Latent Diffusion pre-trained model to generate images of "a turtle in a car with a hat"...That did not quite work:
Image of what is supposed to be a Turtle in a car with a hat, but ultimately its just a garbled mess of colors and shapes
Later on we moved on to using Stable Diffusion, and the results were WAY better! 🐢🚗🤠 Check out the Stream VOD for the results! 👇

Links from today's episode

🐦 Reach out to the hosts and guests:

Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.